Reinforcement Learning on Minepump 1.0 (evaluation)
-669.16RewardDQN + Static 1
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| DQN + Static 1Shielding=Static 12025.11 | -669.16 | 0 | 0.01 | |
| DQNShielding=None2025.11 | -672.36 | 0 | — | |
| DQN + Static (*)Shielding=Static (*)2025.11 | -747 | 0 | 0.4 | |
| DQN + AdaptiveShielding=Adaptive2025.11 | -764.98 | 1 | 0.19 | |
| DQN + Static 2Shielding=Static 22025.11 | -813.9 | 0 | 0.4 | |
| Static Ctrl. 1Type=Symbolic Controller2025.11 | -821.1 | 0 | — | |
| Static Ctrl. 2Type=Symbolic Controller2025.11 | -1,018.29 | 1 | — |